Hi Pritha!
I’m the Product Manager for Qt for Python in The Qt Group. First of all, thank you for your interest in Qt for Python and in PySide as a part of the former! I need to address few general aspects along with my answer.
Speaking of “licensing considerations for using PySide in this setup”: PySide, actually PySide6 in particular, is in no way different from the “regular” Qt in that aspects. This page https://doc.qt.io/qtforpython-6/commercial/index.html in the docs outlines this. This also applies to terms for deployment. For commercial users, we outline this and other terms here https://www.qt.io/terms-conditions/qt-dev-framework-2025-02 in general, where deployment is covered in Exhibit 1 in particular. If you are a FOSS user, you have to comply with the terms of applicable OSS licenses, which are for PySide6 LGPLv3 and GPLv3.
Speaking of more technical topics and the board you use: I googled a bit and found this one https://www.phytec.eu/en/produkte/development-kits/phyboard-pollux-imaging-kit/#technische-details/. This page mentions it uses Yocto. This is already a good start. Yocto is split into meta-layers. Here, initially, there is not much difference from “regular” Qt when it comes to how Qt is integrated there. Our commercial offering is Boot2Qt. It sits on top of the OSS “meta-Qt6” layer, see https://doc.qt.io/Boot2Qt/b2qt-meta-qt6.html. Boot2Qt has included PySide6 since Qt 6.8.2, see https://doc.qt.io/qtforpython-6/tutorials/embedded/boot2qt.html. This page actually mentions that it is applicable to Raspberry Pi. Raspberry Pi is one of our Tier 1 reference targets, see https://doc.qt.io/qt-6/supported-platforms.html#tier-1-reference-targets. iMX8 is also on the same list as “i.MX 8QuadMax MEK”, but I do not know what is different in your edition of iMX8 compared to “i.MX 8QuadMax MEK.” This is something that one of the embedded folks should answer. Generally, NXP is a good partner with Qt, and we know that Qt (including PySide6) is used in a large number of projects with iMX8. So your board should not be far off. If you are a commercial customer, you could also use Qt Support. In the case of OSS, you might have quite a more to take care of aside from your actual project since all boards are still a bit different and you also need to build Python and Qt for Python for your emebedded target and deploy it as a runtime. This is pre-made in Boot2Qt which I mentioned before.
I hope it helps!